Less Common Viral Infections

Human Papillomavirus

Human papillomaviruses (HPV) are the most common cause of sexually transmitted infection in the United States, with approximately 14 million new infections annually, according to the Centers for Disease Control and Prevention (CDC). These DNA viruses are the cause of condyloma acuminata (i.e., genital warts), cervical condylomata, and cervical cancer, but most infections are asymptomatic and resolve spontaneously within 2 years. More than 40 human HPV serotypes have been found to infect the anogenital tract, and at least 12 serotypes are regarded as carcinogenic. Of these, serotypes 16 and 18 account for greater than 70% of cervical cancer worldwide, and serotypes 6 and 11 account for greater than 90% of anogenital warts (condylomata acuminata). The 2011 National Health and Nutrition Study found that greater than 40% of women in the United States were infected with at least one serotype of HPV, and that women in the 20- to 25-year age group were most likely to be infected.

The infant born to a mother with HPV infection is at risk for development of juvenile laryngeal papillomatosis and/or of anogenital warts, primarily because of HPV serotypes 6 and 11. Hajek associated the presence of condyloma acuminatum in a mother at the time of delivery with the subsequent development of laryngeal papilloma in her infant ( Table 30-1 ). Cook and colleagues described a similar association in nine children with laryngeal papilloma. Five of the children who developed laryngeal papilloma when younger than 6 months of age were born to mothers who had condylomata acuminata at the time of delivery, and the mothers of two of the other four had genital warts but did not have them at the time of delivery. Seven (78%) of the nine children with laryngeal papilloma had mothers with condylomata acuminata. Six of the nine children also had skin warts. The expected incidence of condylomata acuminata in women in the population studied by Cook and colleagues was 1.5%. Quick and coworkers also described a strong association between laryngeal papilloma in young children and maternal condylomata. Twenty-one (68%) of the 31 patients with laryngeal papilloma they studied had been born to mothers who had had condylomata.

HPV cannot be isolated by tissue culture, but HPV DNA sequences can be detected in cervical cells. HPV can be detected in epithelial cells that have a normal histologic appearance and from tissue samples of patients whose papillomatous lesions are in remission. The frequency of HPV detection has ranged from 5% to 20% in studies of women of childbearing age, with the highest incidence occurring among younger women. It has been estimated that greater than 70% of sexually active women will contract HPV during their lifetime, and that the frequency of infection is correlated with the number of sexual partners. Pregnancy has not been associated with a higher rate of infection. Most pregnant women with HPV infection do not have a history of genital warts.

Results of studies investigating the frequency of subclinical transmission from mothers to infants vary widely. The results are likely confounded by the method and site of attempted detection, the quantity of maternal virus, the HPV strain(s) with which the mother is infected, and the duration of infant follow-up. Recent studies using nucleic acid techniques to detect virus have found that 20% to 30% of infants born to HPV-infected mothers are HPV positive early in life, but the strains detected in those infants during the first few weeks of life are generally no longer detected by 6 to 12 months. Perinatal transmission through exposure to the virus at delivery has been proposed as the means of transmission, but HPV DNA has also been found in cord blood, placenta, and breast milk of mothers whose infants were positive for the same genotype, and papillomatosis has been described in infants delivered by cesarean section.

Despite the prevalence of genital HPV infection, juvenile laryngeal papillomatosis remains a rare disease. The incidence of recurrent respiratory papillomatosis is variable, depending upon the epidemiology of HPV infection among women of childbearing age in the population. A national surveillance study in Canada determined that between 1994 and 2007 the incidence was 0.24 per 100,000 children. Others have estimated that the incidence is 7 of every 1000 children born to mothers with vaginal condyloma. Because of the high prevalence of asymptomatic HPV infection, the feasibility of preventing the rare cases of laryngeal papillomatosis by considering maternal condyloma acuminatum as an indication for cesarean delivery is questionable.

Treatment of anogenital warts is not optimal, but podophyllum resin or podofilox is often used in older children and adults. Neither has been tested for safety or efficacy in children, and both are contraindicated for use in pregnancy. Laryngeal papillomas recur even after repeated surgical removal. Interferon has been used with some success for treatment of laryngeal papillomas. Although the mainstay of surgical management has traditionally been the carbon dioxide laser, newer surgical techniques have demonstrated efficacy in the management of pediatric patients, including powered instrumentation and the pulse-dye laser. Adjuvant medical therapies used for pediatric recurrent respiratory papillomatosis include topical interferon-α2a, retinoic acid, and indol-3-carbinol diindolylmethane (I3C/DIM). Topical cidofovir has not demonstrated efficacy compared with placebo.

Two HPV vaccines are available for the prevention, but not therapy, of HPV infection. One vaccine is protective against the two most common serotypes associated with cervical cancer (HPV serotypes 16 and 18) as well as the two most common serotypes associated with anogenital warts (HPV serotypes 6 and 11); this vaccine is licensed in the United States for use in females and males 9 to 26 years of age. A second vaccine protects only against HPV serotypes 16 and 18 associated with cervical cancer and is licensed only for females 10 to 25 years of age. Although two approaches, preventive and therapeutic, have been pursued for HPV vaccines, to date only preventive vaccines, which evoke a robust neutralizing antibody response to selected HPV serotype, have been successful. There is evidence from recent epidemiologic studies that implementation of HPV vaccine recommendations have reduced the prevalence of HPV infection. Further research to develop therapeutic vaccines that generate cytotoxic T lymphocytes and destroy HPV-infected neoplastic cells is underway.

Epstein-Barr Virus

Epstein-Barr virus (EBV) is a human herpesvirus that is most familiar as the cause of infectious mononucleosis; however, most women of childbearing age have been infected asymptomatically in childhood. Because EBV cannot be isolated directly in tissue culture, serologic tests are used to detect recent primary or past infection.

Persons infected with EBV form immunoglobulin G (IgG) and IgM antibodies to viral capsid antigens (VCAs) soon after infection. About 80% form antibodies to early antigens (EAs), which usually fall to undetectable levels 6 months after infection. The presence of antibodies to EAs at later times after acute infection may indicate viral reactivation. Antibodies to EBV-associated nuclear antigen (EBNA) develop weeks to months after primary infection in most people and probably persist for life, as do IgG antibodies to VCAs.

Prospective studies using antibodies to EAs as a marker of recent maternal EBV infection have yielded conflicting results. In a group of 719 women evaluated by Icart and Didier, pregnancies resulting in early fetal death, birth of infants with a congenital abnormality, prematurity or intrauterine growth retardation, and deaths or illnesses during the first week of life were more common in women who were EA antibody positive during the first 3 months of pregnancy than in those who were not. Whether these women had a recent primary EBV infection or reactivation of an infection cannot be determined. In contrast, Fleisher and Bolognese found that the frequency of antibodies to EA in pregnant women was 55%, compared with 22% to 32% among nonpregnant adults, but the incidence of low birth weight, neonatal jaundice, or congenital anomalies was not increased among infants of women with anti-EA antibodies. Esklid and colleagues found no association between EBV reactivation after the first trimester of pregnancy and the likelihood of fetal death in a large Norwegian cohort, but in their study, pregnancy was shorter and birth weight was lower for infants born to mothers found to have high levels of IgM antibody against EA during pregnancy. The rate of congenital anomalies was not determined. Avgil and colleagues found no difference in pregnancy outcome, including miscarriage, prematurity, birth weight, or congenital anomalies for 126 women who had serologic evidence of primary or reactivated EBV infection during pregnancy, compared with 1434 control women. Primary EBV infection during pregnancy is unusual because only 3.0% to 3.4% of pregnant women are susceptible. Recent primary EBV infection is diagnosed by the presence of VCA IgG and IgM antibodies in the absence of antibodies to EBV-associated nuclear antigen. Among six women who had primary EBV infections during pregnancy, as established by the presence of IgM antibody to VCA and the absence of antibody to EBNA in their sera, only one had symptoms compatible with mononucleosis during pregnancy; she gave birth to a normal infant. Four of the remaining five pregnancies terminated abnormally. One woman had a spontaneous abortion, and the other three were delivered of premature infants. All three of the premature infants were abnormal. One was stillborn, one had multiple congenital anomalies, and one was small for gestational age. The products of abortion and the premature infants were not studied for evidence of an EBV infection. The abnormal infants in this study did not have a characteristic syndrome but instead had a variety of abnormalities.

Fleisher and Bolognese identified three infants born to women who had had silent EBV seroconversion during the first trimester. Two infants were normal; one infant had tricuspid atresia. EBV IgM was not detected in cord blood serum, and EBV was not recovered from the cord blood lymphocytes. Three infants of mothers with a primary EBV infection and infectious mononucleosis were normal at birth and had no serologic or virologic evidence of intrauterine infection.

Early reports implicated EBV as a cause of congenital anomalies, particularly congenital heart disease; however, Tallqvist and colleagues were unable to detect an increase in incidence of antibodies to EBV in 6- to 23-month-old children with congenital heart disease compared with normal, age-matched controls. EBV may cause congenital heart disease in an individual case, but this study suggests that it is not a common cause of cardiac defects. Goldberg and associates described an infant born with hypotonia, micrognathia, bilateral cataracts, metaphyseal lucencies, and thrombocytopenia. Immunologic evidence suggesting possible EBV infection included an elevated total IgM level, the presence of IgM anti-VCA antibody at 22 days of age, and a delay in development of anti-EBNA antibody until 42 days of age. Weaver and coworkers described an infant with extrahepatic bile duct atresia and evidence of intrauterine EBV infection; EBV IgM was identified in serum obtained when the infant was 3 and 6 weeks of age, and persistent EBV IgG was seen at 1 year.

Although EBV cannot be recovered by standard tissue culture methods, the virus can be detected by its capacity to transform B lymphocytes into persistent lymphoblastoid cell lines. In studies conducted to identify cases of intrauterine EBV infection, Visintine and colleagues and Chang and Blankenship observed spontaneous transformation of lymphocytes obtained from cord blood, but EBV could not be associated. EBV-transformed cells were not found in any samples of cord blood from 2000 newborns studied by Chang and Seto or from the 25 newborns tested by Joncas and associates. One study used nested polymerase chain reaction (PCR) methods for amplifying EBV DNA regions in circulating lymphocytes from 67 mother-infant pairs within 1 week of birth. Approximately 50% of the women and two of the neonates were EBV PCR positive. Visintine and colleagues studied 82 normal-term infants, 28 infants with congenital anomalies, and 29 infants suspected of having congenital infections; they were unable to isolate EBV from any of these infants. Two infants have been described in whom there was evidence of infection with EBV at birth. A congenital cytomegalovirus (CMV) infection coexisted in both. Most of the clinical findings in the infants were compatible with those typical of congenital CMV infections and included microcephaly, periventricular calcifications, hepatosplenomegaly, and inclusions characteristic of CMV in sections of tissues or cells in urinary sediments. One infant had deformities of the hands similar to those seen in arthrogryposis. Neither CMV nor EBV was isolated from the saliva or secretions of these infants. In the first infant, IgM antibody to EBV was present at birth and EBNA-positive permanent lymphoblastoid cell lines were established on five occasions between 3 and 30 months of age. In the second infant, permanent lymphoblastoid cell lines were established from the peripheral blood at birth and from postmortem heart blood at 3 days of age. EBNA and EBV RNA were identified in these cells, and CMV DNA was identified in the cells from the liver of the same infant.

Attempts to isolate EBV from secretions obtained from the maternal cervix have been unsuccessful, but the virus can be detected at this site by DNA hybridization. There is little evidence suggesting that natal transmission of EBV occurs. However, EBV was recovered from genital ulcers in a young woman with infectious mononucleosis. Fatal EBV infection was diagnosed by DNA hybridization of lymph node tissue from one infant who presented with failure to thrive, emesis, diarrhea, and a macular rash at 14 days of age, but this infection might have been acquired in utero.

EBV can be transmitted to newborns in the perinatal period by blood transfusion. Permanent lymphoblastoid lines that contained EBV antigens were established by Joncas and coworkers from the blood of two infants who had been transfused. One of these infants did not develop permanent antibodies to EBV.

There is no convincing evidence at present that EBV causes congenital anomalies. Because the early and the late serologic responses of young infants to a primary EBV infection differ from those found when a primary infection occurs at an older age, it will be difficult to screen large numbers of newborns for serologic evidence of an EBV infection sustained in utero.

Human Herpesvirus 6

Human herpesvirus 6 (HHV-6) is a member of the Betaherpesvirinae subfamily of herpesviruses; HHV-6 has been identified as a cause of exanthema subitum (i.e., roseola). The virus exhibits tropism for T lymphocytes and is most closely related to human CMV by genetic analysis.

Seroepidemiologic studies have shown that HHV-6 is ubiquitous in the human population, regardless of geographic area, and that it infects greater than 90% of infants during the first year of life. IgG antibodies to HHV-6 are detected in almost all infants at birth, with a subsequent decline in seropositivity rates by 4 to 6 months of age as transplacentally acquired antibody is lost. The highest rate of acquisition of HHV-6 infection appears to occur during the first 6 months to 1 year of life as maternal antibodies wane.

The seroepidemiologic evidence and restriction enzyme analysis of paired virus isolates from mothers and their infants suggest that the usual route of transmission is perinatal or postnatal. HHV-6 is present in the cord bloods of approximately 1% of newborns , and HHV-6 genome was found in the amniotic fluid of a similar proportion (7/729) of pregnancies. Hall and colleagues have demonstrated that the majority of congenitally acquired infection is transmitted as chromosomally integrated HHV-6 transmitted from a parent, and in those cases, HHV-6 DNA can be demonstrated in all tissues tested. Among 43 infants with congenital infection, they found evidence of chromosomal integration in 86%, whereas the remaining 14% had acquired their infection transplacentally. Cord blood anti–HHV-6 antibodies were similar whether the infection was chromosomally integrated or transplacentally acquired.

Primary HHV-6 infection should be rare during pregnancy because almost all adult women have been infected in childhood. One case of a spontaneous abortion has been described in a woman who was seronegative to HHV-6 at the beginning of pregnancy and had developed anti-HHV-6 antibody by the time of the pregnancy loss. HHV-6 was detected by PCR assay in fetal tissues, the placenta, umbilical cord blood, and maternal serum. Symptoms attributable to congenital HHV-6 infection have been only rarely described in live-born infants. Lanari and colleagues described a term newborn who developed seizures during the first few hours of life and was found to have HHV-6 DNA in the cerebrospinal fluid (CSF) on day 2. The infant developed evidence of neurologic compromise with developmental delay. Evidence of reinfection after presumed congenital HHV-6 infection also has been demonstrated.

In addition to the roseola syndrome, HHV-6 has been detected by PCR in peripheral blood lymphocytes obtained from infants younger than 3 months who had acute, nonspecific, febrile illnesses. Two neonates who had fulminant hepatitis associated with HHV-6 infection have been described. Other associations found among infants include a mononucleosis-like syndrome, pneumonitis, and one case report of possible immunodeficiency and pneumonitis associated with HHV-6 infection. However, all clinical associations between disease in infants and HHV-6 infection must be evaluated with care because of the evidence that most infants become infected with this virus within a few months after birth and that the virus persists after primary infection, as is characteristic of herpesviruses. In an attempt to identify clinical symptoms related to HHV-6 infection, Rentz and colleagues tested infants hospitalized in their neonatal intensive care unit for HHV-6 and compared clinical findings in 12 infants whose plasma was positive for HHV-6 DNA with 144 HHV-6–negative infants. In this small study, no significant differences were found in clinical characteristics of the two groups.

Human Herpesvirus 7

Human herpesvirus 7 (HHV-7) was discovered in the peripheral blood lymphocytes of a healthy adult in 1990. Like HHV-6, HHV-7 belongs to the Roseolovirus genus within the Betaherpesvirinae subfamily, along with HHV-6 and CMV. Like HHV-6, it causes primary infection in most individuals during childhood. However, clinically symptomatic infection with HHV-7 appears to be significantly less common and occurs later in life than HHV-6.

The average age at infection is about 2 years, and 75% of children are seropositive by 5 years of age. The primary mechanism of transmission is from contact with saliva of infected individuals. Because HHV-7 DNA has been detected in breast milk, breastfeeding may be another source of infection. However, antibodies to HHV-7 in breast milk may protect against infection, and in one study, breastfeeding was associated with a lower risk of early acquisition of HHV-7 infection. HHV-7 DNA has been detected in 2.7% of cervical swabs obtained from women in their third trimester of pregnancy but from none of the swabs of control women, suggesting that pregnancy may be associated with reactivation of HHV-7. However, perinatal transmission from contact with infected maternal secretions is unknown, and neonatal infections with HHV-7 have not been reported. Clinical symptoms are rarely associated with HHV-7 infection but include nonspecific fever, with or without rash, which resembles exanthema subitum. Clinically apparent HHV-7 infections appear to have a high rate of central nervous system (CNS) involvement.

Influenza A and B

Early investigations of the teratogenic potential of influenza virus were epidemiologic studies in which the diagnosis of influenza was not confirmed serologically. Some of those studies suggested that infants born to mothers who experienced clinical influenza infection during pregnancy were more likely to be born with congenital anomalies, particularly CNS anomalies, including anencephaly. Record and Leck analyzed previously collected data and were unable to find an association between influenza and malformations of the CNS. An increase in congenital defects in infants of mothers who had influenza-like symptoms at 5 to 11 weeks of gestation was reported by Hakosalo and Saxen. Most of these anomalies involved the CNS, but there was no increase in incidence of anencephaly in infants of women who had symptoms compatible with influenza compared with those who remained asymptomatic.

All of these studies were undertaken during influenza epidemics. It was assumed that, under these circumstances, there would be a high correlation between a history of influenza, as elicited from the patient, and infection with influenza virus. MacKenzie and Houghton summarized the reports implicating influenza virus as a cause of maternal morbidity and congenital anomalies and came to the conclusion that probably no association exists between maternal influenza infection and subsequent congenital malformations or neoplasms in childhood.

Several studies have been performed in which influenza infection during pregnancy has been serologically confirmed. Hardy and coworkers reported that the incidence of stillbirths was higher in 332 symptomatic pregnant women with serologically confirmed influenza infections than in 206 women with serologically confirmed infections who had remained asymptomatic or in 73 uninfected women. The control group of uninfected women was smaller than expected because the attack rate during the period of the study was very high. Major congenital anomalies occurred in 5.3% of women whose infections occurred during the first trimester compared with 2.1% of 183 women infected during the second trimester and 1.1% of 275 women infected during the third trimester. Supernumerary digits, syndactyly, and skin anomalies were excluded from these figures. Among infants of mothers infected during the first trimester, cardiac anomalies were the most common type of abnormality; none of these infants had anencephaly. Griffiths and associates observed a slight increase in congenital anomalies in infants born to women who had had serologically confirmed influenza during pregnancy compared with infants of women who had not; however, all of the infants with congenital anomalies were born to women who had had influenza in the second or third trimester, suggesting that congenital infection was an unlikely cause.

Influenza infections during pregnancy are more likely to result in hospitalization for respiratory symptoms in the pregnant woman than in nonpregnant adults. Hartert and associates conducted a matched cohort study of pregnant women to determine pregnancy outcomes associated with respiratory hospitalizations during influenza seasons from 1985 to 1993. During those influenza seasons, 293 pregnant women were hospitalized for respiratory symptoms at a rate of 5.1 per 1000 pregnant women. The prevalence of prematurity and low birth weight was not higher than a matched cohort of pregnant women hospitalized with nonrespiratory diagnoses. However, pregnant women with asthma had higher rates of respiratory hospitalizations than those without asthma, and all of three fetal deaths in this cohort were singleton, late third-trimester, intrauterine fetal deaths in mothers who had asthma and were current smokers. During the 2009 H1N1 influenza pandemic, the CDC requested state and local health departments to report all pregnant women hospitalized with influenza. Over an 8.5-month period in 2009, the CDC received reports of 347 severely ill women, 272 of whom were admitted to intensive care units but survived, and 75 of whom died. Preterm delivery was more common for the women who were severely ill and delivered during their hospitalization, compared with the general population, and infants delivered following their mother’s hospitalization for influenza infection were more likely to be born small for gestational age. In this group of very sick women, antiviral therapy was provided earlier for the women who survived. A report from Great Britain of 256 pregnant women hospitalized during the 2009 pandemic described an increase in perinatal mortality compared with the general population, primarily because of an increase in stillbirths.

It can be said with certainty that intrauterine exposure to influenza virus does not cause a consistent syndrome. If there is a cause-and-effect association between influenza virus infections during pregnancy and congenital anomalies, the latter occur with low frequency. Hakosalo and Saxen have documented an increase in the use of nonprescription drugs during influenza outbreaks and have suggested that drugs rather than infection with influenza virus may exert an erratic teratogenic influence. A number of studies have investigated the possible association between influenza infection in pregnant women and subsequent development of bipolar affective disorders or schizophrenia among their offspring, with mixed results.

Viremia is rare during influenza infections, but it does occur. Few attempts have been made to demonstrate transplacental passage of the virus to the fetus. Ruben and colleagues tested the cord sera of infants born to 22 mothers who had been pregnant during an influenza A/England/42/72 outbreak and had serologic evidence of infection. All seven infants with evidence of antigenic recognition of influenza virus at birth had uncomplicated deliveries and remained healthy. Influenza A/Bangkok was isolated from the amniotic fluid of a mother with amnionitis and acute influenza infection at 36 weeks of gestation; the infant who was born at 39 weeks of gestation had serologic evidence of infection but was asymptomatic.

Yawn and associates studied a woman who developed influenza in the third trimester and died of pulmonary edema. A virus similar to the prototype strain A ₂ /Hong Kong/8/68 was isolated from the lung, hilar nodes, heart, spleen, liver, kidney, brain, and spinal cord of the mother and from the amniotic fluid and myocardium of the fetus, which did not survive. Ramphal and colleagues studied another woman who died of complications of an influenza infection at term. A virus similar to strain A/Texas/77 was isolated from maternal tissues, but influenza virus was not isolated from any of the fetal tissues tested.

In contrast to intrauterine infections with influenza virus, which are rare, infections acquired by infants in the neonatal period are common. Passively transferred antibody to influenza virus may prevent symptomatic infections during the first few months of life if it is present in sufficient quantity. Outbreaks of influenza virus infection have occurred in neonatal intensive care units. Although some newborns found to be infected during outbreaks remain asymptomatic, others have a variety of respiratory symptoms, including pneumonia. Influenza virus infection may be fatal in the neonatal period. In December 2012, the U.S. Food and Drug Administration (FDA) approved the use of oseltamivir for treatment of infants as young as 2 weeks of age. Safety of this drug during the neonatal period had been demonstrated in studies performed during the 2009 H1N1 influenza pandemic. Oseltamivir is the only antiviral agent approved for use in newborns, and it is not approved for prophylactic use.

Infants younger than 6 months cannot be protected by influenza vaccine. All health care professionals should receive influenza vaccine annually in the fall. Several studies have demonstrated that immunization of pregnant women provides protection for their newborns exposed to influenza. Since 2004, the CDC has recommended that all pregnant women receive influenza vaccine during pregnancy, regardless of the trimester in which it is available.